`git clone -h` does not contain "--depth"

Hi, In build-aux/bootstrap there is this fragment of code: if git clone -h 2>&1 | grep -- --depth > /dev/null; then shallow='--depth 2' fi But the output of `git clone -h` nowadays (version 2.48.1) does not contain the option --depth. It instead contains this: --[no-]dept

[PATCH] terminfo, termcap: Fix more spellos in a comment.

-27 Benno Schulenberg(tiny change) + + terminfo, termcap: Fix more spellos in a comment. + * lib/tparm.c: Fix misspellings and wording in the main comment. + 2023-02-27 ChuanGang Jiang fts: fail gracefully when out of memory diff --git a/lib/tparm.c b/lib/tparm.c index

glob.m4 leaves the file 'conf-file' behind

Since commit c386ed6eb0 from a month ago, the m4/glob.m4 script creates a file './conf-file' and does not delete it afterward, causing 'make distcheck' to fail for GNU nano. ERROR: files left in build directory after distclean: ./conf-file Makefile:2038: recipe for target 'distcleancheck' failed

Re: build failure of glob module on NetBSD: request for member ‘dd_fd’...

Op 27-04-2022 om 00:02 schreef Paul Eggert: > I took the easy way out, though, and simply documented the bugs and modified > glob to not run afoul of the bugs, by installing the attached patch into > Gnulib. This fixes the issue for me. Thanks. Benno OpenPGP_signature Description: OpenPGP d

build failure of glob module on NetBSD: request for member ‘dd_fd’...

Trying to build nano from git on a NetBSD 9.2 virtual machine, I get this: [...] gcc -DHAVE_CONFIG_H -I. -I.. -D_NETBSD_SOURCE -D_XOPEN_SOURCE=600 -I/usr/pkg/include/ncursesw -I/usr/pkg/include -I/usr/pkg/include -Wno-cast-qual -Wno-conversion -Wno-float-equal -Wno-sign-compare -Wno-undef -W

[PATCH] doc: add two missing closing parentheses

100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +2022-02-22 Benno Schulenberg(tiny change) + + doc: add two missing closing parentheses + * doc/regex.texi (Syntax Bits): Add missing closing parenthesis. + * doc/regex.texi (BSD Regular Expression Compiling): Likewise

Re: compiling with tcc fails on regex.c since commit 56b32251

Op 27-11-2021 om 10:01 schreef Paul Eggert: > tcc is failing because it claims to support C99 (it > defines __STDC_VERSION__ to 199901L), but it doesn't support variable length > arrays as C99 requires. Perhaps you could file a bug report with the tcc > folks? Reported as https://savannah.nongnu

compiling with tcc fails on regex.c since commit 56b32251

Hi, Since commit 56b32251 from two months ago, I can no longer compile nano with tcc. It breaks on the regex module: [...] source='regex.c' object='regex.o' libtool=no \ DEPDIR=.deps depmode=tcc /bin/bash ../depcomp \ tcc -DHAVE_CONFIG_H -I. -I.. -D_GNU_SOURCE -D_DEFAULT_SOURCE -I/usr/include

Re: heap-buffer overflow when searching for regex @\*

Op 19-10-2021 om 00:10 schreef Paul Eggert: > Thanks for the bug report. I reproduced the problem with texinfo and installed > the attached Gnulib patch, which fixed things for me with texinfo. I didn't > quite > follow the 'nano' instructions but I hope this fixes the problem there too. Yes, us

heap-buffer overflow when searching for regex @\*

Hi, When compiling the 'info' program or GNU nano with -fsanitize=address, then searching in either of the programs for the regex "@\*" (without the quotes) causes an abortion in gnulib's re_search_internal() at lib/regexec.c:764. To reproduce, configure texinfo-6.8 with CFLAGS="-g -O0 -march=na

mkstemps is not listed alphabetically in the manual

Hi, On https://www.gnu.org/software/gnulib/manual/html_node/Glibc-stdlib_002eh.html the entry for 'mkstemps' is out of alphabetical order. This is confusing. (I had downloaded the PDF of the manual, searched for mkstemps, paged up a bit, and then -- scrolling back down -- wondered where mkstemp

Re: commit eb32c992 causes trouble on older Debian

Op 27-04-2021 om 20:28 schreef Paul Eggert: > Thanks for reporting that. I installed the attached into Gnulib to try to work > around the problem. Please give it a try (I haven't tested it on older > Debian). Confirmed. Commit afeeb210 fixes the problem. Thanks. Benno signature.asc Descrip

commit eb32c992 causes trouble on older Debian

Hi, Commit eb32c992 from five days ago causes trouble when trying to compile GNU nano from git on an older Debian box: make[4]: Entering directory `/home/bens/NANO/lib' depbase=`echo malloc/dynarray_at_failure.o | sed 's|[^/]*$|.deps/&|;s|\.o$||'`;\ gcc -std=gnu11 -DHAVE_CONFIG_H -I. -I.

Re: [PATCH] mbrtowc: remove a redundant condition

Op 21-03-2021 om 19:22 schreef Bruno Haible: > Hi Benno, > >> * lib/mbrtowc-impl-utf8.h: There is no need to check for c == 0xf4 >> when !(c < 0xf4), as ten lines earlier c <= 0xf4 was established. > > Your patch is correct. But we generally don't need to do micro- > optimizations that a compile

[PATCH] mbrtowc: remove a redundant condition

index 76ea9c3eb..d2dfe5e0c 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +2021-03-21 Benno Schulenberg(tiny change) + + mbrtowc: remove a redundant condition + * lib/mbrtowc-impl-utf8.h: There is no need to check for c == 0xf4 + when !(c < 0xf4), as ten lines

Re: commit 452fee36 causes trouble when compiling with clang

Hello Bruno, >> In file included from regex.c:74: >> ./regexec.c:1198:3: error: invalid application of 'typeof' to bit-field >> DEBUG_ASSERT (state->halt); >> ^ > > Thanks for the report. I didn't know that __typeof__ does not work for all > kinds of expressions. (intprops uses __typeof__ as

commit 452fee36 causes trouble when compiling with clang

Hi, Running a compile test of nano with the current state of gnulib (commit 452fee36, "verify: Do use __builtin_assume on clang") on FreeBSD (which happens to use clang), the compile failed: clang -DHAVE_CONFIG_H -I. -I.. -D_GNU_SOURCE -D_DEFAULT_SOURCE -I/usr/include/ncursesw -g -O1 -march=

[PATCH] users.txt: add GNU nano

Nano has been making use of gnulib since March 2017, version 2.8.0. --- ChangeLog | 5 + users.txt | 1 + 2 files changed, 6 insertions(+) diff --git a/ChangeLog b/ChangeLog index 25f1f3e3b..983dcf4c8 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,8 @@ +2019-10-06 Benno Schulenberg

Re: gnulib's translation

Op 02-12-18 om 13:10 schreef Bruno Haible: > - Thus the interface between the package maintainers and the TP is simpler > if > the package maintainers submit a tarball to the TP. Precisely. Furthermore, when following git, how is the translator to know that a release is approaching and t

Re: gnulib's translation

Op 01-12-18 om 17:00 schreef Bruno Haible: > Akim has just moved some code from Bison to Gnulib. He writes: > >> Also, I feel sorry for Bison's translators when submitting modules to >> gnulib: it's kind of throwing away their work; it would be great if there >> were a means to preserve these tr

Re: gnulib's translation

> Akim Demaille wrote: >> maybe the translation project could work on top of git now? No, I'm not going to do that. The TP is geared toward packages that make releases. So I need a (prerelease) tarball that contains the corresponding POT file. Benno

[PATCH 2/2] bootstrap, gnulib-tool: use https instead of insecure rsync

The rsync command does not do any authentication and thus allows man-in-the-middle attacks. Better use wget over https, although this is slower. * build-aux/bootstrap (download_po_files, po_download_command_format): Don't try using rsync; always use wget over https to fetch PO files. * gnulib-too

[PATCH 1/2] bootstrap, gnulib-tool: correct the translations wget command

Using the -np (--no-parent) option doesn't prevent wget from traversing the given URL to a default depth of 5, and since the domain directory contains a href="/latest/" link, this means that wget searches through all domains anyway and will also download PO files that are meant for other packages.

Re: [PATCH V2] localcharset: short-circuit the search for an alias on a Mac

Hi Bruno, Op 21-04-18 om 12:41 schreef Bruno Haible: > Thanks, applied. (For some background: a bug reporter found that on a Mac lots of time would be spent in locale_charset() when nano called wcwidth(): https://savannah.gnu.org/bugs/?53562#comment16.) > I added a comment. > Explanations about

[PATCH V2] localcharset: short-circuit the search for an alias on a Mac

the list is avoided. --- ChangeLog | 10 ++ lib/localcharset.c | 3 ++- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/ChangeLog b/ChangeLog index f5752a217..1c6f32294 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,13 @@ +2018-04-21 Benno Schulenberg(tiny

[PATCH] localcharset: short-circuit the search for an alias on a Mac

the list is avoided. --- ChangeLog | 10 ++ lib/localcharset.c | 3 ++- 2 files changed, 12 insertions(+), 1 deletion(-) diff --git a/ChangeLog b/ChangeLog index f5752a217..1c6f32294 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,13 @@ +2018-04-21 Benno Schulenberg(tiny

[PATCH] stat: fix compilation failure on macOS Sierra

Reported by Marius Schamschula in https://savannah.gnu.org/bugs/?52546. * lib/stat.c: Add missing include of stat-time.h. --- lib/stat.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/lib/stat.c b/lib/stat.c index 30ba67efc..f52ae9222 100644 --- a/lib/stat.c +++ b/lib/stat.c @@ -47,6 +47,

Re: clang's Undefined Sanitizer

Op 24-08-2017 om 11:34 schreef Bruno Haible: This part of the patch: [...] discards your change http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=e29357c8f445320cb3084352579fb2648cce4d0f I haven't gotten any answer to my inquiries https://lists.gnu.org/archive/html/bug-gnulib

Re: clang's Undefined Sanitizer

Hi Bruno, Op 24-08-2017 om 11:34 schreef Bruno Haible: [...] discards your change http://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=e29357c8f445320cb3084352579fb2648cce4d0f I haven't gotten any answer to my inquiries https://lists.gnu.org/archive/html/bug-gnulib/2017-07/msg9

[patch] glob: do not place declarations after code, so it will compile on Haiku

, I need only the first hunk, but since a comment on the next function says to keep the two functions in sync...) (Please CC; not subscribed.) Benno >From 63c8fb5f56210989ba33230c77febbc3323aafa0 Mon Sep 17 00:00:00 2001 From: Benno Schulenberg Date: Sat, 1 Jul 2017 18:05:00 +0200 Subject: [PA

why does the gnupload script itself ask for the passphrase?

Hi, When using the gnupload script to upload a release tarball, it first asks (on the terminal): "Enter GPG passphrase:". I type the passphrase, it says: "Signing nano-1.2.3.tar.xz ...", and then the graphical gpg-agent pops up and asks for the passphrase again. What gives? It seems I could put

[patch] fix the grammar of some comment lines

Hi, Looking at propernam(), I noticed a few tiny grammar mistakes in the comments. Attached patch fixes those. Benno -- http://www.fastmail.com - Access your email from home and the web From f4637c6a3dc674fb97c570f57aec06680551183d Mon Sep 17 00:00:00 2001 From: Benno Schulenberg Date: Sun

Re: Vietnamese .po file causes coreutils build failure [Re: bootstrap: TP urls changed

Jim Meyering wrote: > You can argue that the translation project should not publish > such a .po file, and I agree that "upstream" (in the TP > .po-checking code) is the right place to fix this, but can you > get them to fix it quickly I've run the following on all files in the latest/ directory t